Premium
The influence of the tropical and subtropical Atlantic and Pacific Oceans on precipitation variability over Southern Central South America on seasonal time scales
Author(s) -
Berri Guillermo J.,
Bertossa Germán I.
Publication year - 2004
Publication title -
international journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0899-8418
DOI - 10.1002/joc.1000
Subject(s) - empirical orthogonal functions , climatology , precipitation , subtropics , sea surface temperature , environmental science , tropical atlantic , subtropical ridge , common spatial pattern , canonical correlation , geography , geology , meteorology , statistics , mathematics , artificial intelligence , fishery , computer science , biology
This paper studies the temporal and spatial patterns of precipitation anomalies over southern central South America (SCSA; 22–40°S and 54–70°W), and their relationship with the sea‐surface temperature (SST) variability over the surrounding tropical and subtropical Atlantic and Pacific Oceans. The data include monthly precipitation from 68 weather stations in central–northern Argentina and neighbouring Brazil, Paraguay and Uruguay, and monthly SSTs from the NOAA dataset with a 2° resolution for the period 1961–93. We use the method of canonical correlation analysis (CCA) to study the simultaneous relationship between bi‐monthly precipitation and SST variability. Before applying the CCA procedure, standardized anomalies are calculated and a prefiltering is applied by means of an empirical orthogonal function (EOF) analysis. Thus, the CCA input consists of 10 EOF modes of SST and between 9 and 11 modes for precipitation and their corresponding principal components, which are the minimum number of modes necessary to explain at least 80% of the variance of the corresponding field. The results show that November–December presents the most robust association between the SST and SCSA precipitation variability, especially in northeastern Argentina and southern Brazil, followed by March–April and May–June. The period January–February, in contrast, displays a weak relationship with the oceans and represents a temporal minimum of oceanic influence during the summer semester. Based on the CCA maps, we identify the different oceanic and SCSA regions, the regional averages of SST and precipitation are calculated, and linear correlation analysis are conducted. The periods with greater association between the oceans and SCSA precipitation are November–December and May–June. During November–December, every selected region over SCSA reflects the influence of several oceanic regions, whereas during May–June only a few regions show a direct association with the oceans. The Pacific Ocean regions have a greater influence and are more widespread over SCSA; the Atlantic Ocean regions have an influence only over the northwestern and the southeastern parts of SCSA. In general, the relationship with the equatorial and tropical Atlantic and Pacific Oceans is of the type warm–wet/cold–dry, whereas the subtropical regions of both oceans show the opposite relationship, i.e. warm–dry/cold–wet. Copyright © 2004 Royal Meteorological Society